Our work focuses on two major research themes. One major theme is to elucidate normative mechanisms of fear and anxiety in healthy individuals and to identify which of these mechanisms are perturbed in anxiety disorders. Anxiety is an adaptive response to threat that enables organisms to rapidly and efficiently confront challenges. When anxiety is evoked by anticipation of an unpleasant stimulus such as a loud sound or a mild electric shock (induced-anxiety), our sensorimotor system becomes primed to detect and react quickly to any threat. Our past work shows that this effect is mediated by increased communication between an aversive amplification circuit, that is, the dorsal-medial prefrontal cortex (dmPFC) and the amygdala. Attentional and behavioral biases for threat play a central role in the etiology and maintenance of anxiety disorders. We have now shown that this dmPFC-amygdala aversive amplification circuit remains constantly activated in patients with anxiety disorders, even in the absence of an external stressor. These results indicate that in pathological anxiety, an adaptive threat detection mechanism is chronically switched-on, even in innocuous contexts, and contributes to a crippling focus upon negative life events. We have shown that this aversive amplification circuit is inhibited by serotonin, suggesting that anxious patients may have a weakened serotonin-mediated inhibition. In fact, serotonin plays a key role in the treatment of anxiety disorders. Indeed, a major psychopharmacological treatment of anxiety disorders (i.e., selective serotonin reuptake inhibitor or SSRI such as Prozac) acts by increasing serotonin function. This raises the possibility that the therapeutic effect of SSRI is mediated by reestablishing the serotonin inhibition of communication between the dmPFC and the amygdala. Additionally, we have identified a potential vulnerability marker for anxiety disorders. Anxiety disorders are conceptualized as a disposition to overreact with fear and anxiety to mildly threatening stimuli. However, fear and anxiety are distinct emotional states. Presenting predictable and unpredictable threat is a means of inducing fear and anxiety, respectively. A key finding is that patients with anxiety disorders, including panic disorder, posttraumatic stress disorder, and general anxiety disorder, have a propensity to show elevated anxiety to unpredictable threat. However, our recent results suggest that this propensity characterizes only a subgroup of patients who have a history of panic attack. In addition, we found that the anxiety response to unpredictable threat is heritable. Thus, an exaggerated response to unpredictable threat may constitute a risk factor for panic attack and related anxiety disorders. We also began to map the neural responses to predictable and unpredictable threat, particularly within the regions that have been implicated in animal models, such as the amygdala and the bed nucleus of the stria terminalis (BNST). A current objective is to better delineate the amygdala-BNST circuitry and its connection with the distributed network mediating aversive responses to threat, which includes the insula, dmPFC, and hippocampus. To this goal, we are moving from the traditional 3 Tesla to the state-of-the-art high-resolution 7 Tesla MRI scanners, with which we are studying healthy individuals and patients with anxiety disorders. A second theme concerns the nature of the interactions between anxiety and task performance. On the one hand, it is clear that along with the emotional facets of anxiety disorders, anxiety patients report difficulty concentrating together with feeling distracted, which in turn can negatively impact their job performance and interpersonal relationships. On the other hand, it is also clear that humans possess the capacity to actively disengage from anticipation of a potential threat in order to focus on a challenging task, thereby reducing their anxiety. This resource shift is adaptive because it allows individuals to perform necessary procedures in order to survive in an environment with imminent threat (e.g., war). We have examined these interactive processes using induced-anxiety during performance of working memory (WM) tasks. WM is central to healthy functioning because it supports online maintenance and manipulation of information (e.g., tallying the cost of a grocery bill while shopping). We showed that the impact of anxiety on verbal WM performance, and, reciprocally, of task demand on anxiety is based on the cognitive demand of a given task. Specifically, when a WM task is easy, anxiety remains relatively high and performance is disrupted. However, when a task places greater demand on cognitive resources, anxiety is reduced, and performance normalizes. The dmPFC is a key structure where competition to process threat-related and WM-related information takes place. Further insight into these mechanisms reveals that anxiety and task demand increase communication between dmPFC and amygdala and dorsolateral PFC (dlPFC), respectively. Under low-load (easy tasks), anxiety engages the dmPFC, and dmPFC-amygdala coupling increases to promote an anxious response. Under high-load (difficult tasks) WM engages the dmPFC and dmPFC-dlPFC coupling increases to promote focus on task demands. By describing the neural underpinnings of anxiety/cognition interactions we are one step closer to identifying biomarkers of vulnerability to anxious pathology, and discrete neural sites for novel treatments that target both emotional and cognitive disruptions associated with anxiety.